Mavic 2 Pro Crash Dive

[billt1970]

Friends,

First, I'm an experienced certified commercial remote pilot. Thursday, 10/7/20, I was flying a simple mission (at 84 Jones Station Rd, Arnold MD, 21012) to capture a few photos for a real estate listing that was difficult to capture from the ground. All went well for 4+ minutes during which I captured several decent photos for my client, mostly from the 400' ceiling. Then everything went crazy! I was watching the screen to compose the next shot when I realized that the drone was sliding laterally. I attempted to regain control, but there was no response. I next noticed that the altitude was rapidly decreasing and as it neared the ground I heard the sound of it crashing through a nearby tree and landing with a thud. Remarkably, there was still a camera image on the iPhone with the controller! I quickly located the aircraft and immediately realized that there were no propellers remaining! It's now in for repair for the tune of $785, mostly from a destroyed gimbal. We'll see if insurance covers any of it. In the meantime, I would definitely like to know what happened. One hypothesis is a thrown propeller that led to the catastrophic landing.

I'm attaching the photos I took from the air; a photo of the drone as it landed and the flight records and flight log.

Thanks in advance for any information on what caused the crash, and anything I can do in the future to preclude a similar event.

Best Regards,

Bill Taylor
Gambrills, MD

 

[sar104]

That's unusual. It looks most like an IMU problem, but a definite determination will require the mobile device DAT file ending FLY012.DAT, so I hope that you either retrieved it already or haven't synchronized your flight records since the flight.

The problem started at 231.5 seconds and the FC completely lost control of the aircraft.



And the IMU thought that the aircraft was ascending, when it was descending - implying accelerometer failure and hence my suspicion that the IMU failed.

 

[Prismatic]

Because @sar104 didn't include and motor speed data, I'll assume they were nominal throughout. That rules out in-flight loss of propulsion (not that anyone's ever lost all props in one shot!). But if the props came off passing through the branches, you'd expect some evidence of such violence. Were there any stubs or even hubs still attached? I'm not that familiar with the M2s, but I would expect that tearing off a prop that way--let alone four--would leave some kind of mark.

And, hey, bummer. Sorry to hear of it. I guess it could have ended worse, so there's that small consolation. ?

 

[slup]

Because @sar104 didn't include and motor speed data, I'll assume they were nominal throughout. That rules out in-flight loss of propulsion (not that anyone's ever lost all props in one shot!). But if the props came off passing through the branches, you'd expect some evidence of such violence. Were there any stubs or even hubs still attached? I'm not that familiar with the M2s, but I would expect that tearing off a prop that way--let alone four--would leave some kind of mark.

And, hey, bummer. Sorry to hear of it. I guess it could have ended worse, so there's that small consolation. ?

The motor data isn't available yet as the DAT log attached by OP aren't the correct one ... but agree with @sar104, this is most likely due to a IMU failure, pretty clear seeing a decreasing height together with a negative z-speed (meaning ascending) ... the big difference in IMU & GPS velocities indicate the same. Let's see what we can pinpoint once the correct mobile device DAT log is available.

 

[billt1970]

Thanks for the quick response and analysis, @sar104 ! Unfortunately I did sync the flight records, and so the FLY012.DAT information is lost. I won't sync in the future until I have retrieved that information.

The only recent unusual activity was last Monday, 5/4/20, when I flew in extremely high winds, later documented in the 30-40kph range. Could that have caused some problems that showed up three days later? I have the flight records and logs if that might be helpful. Just let me know.

Best Regards,

BT

 

[billt1970]

The motor data isn't available yet as the DAT log attached by OP aren't the correct one ... but agree with @sar104, this is most likely due to a IMU failure, pretty clear seeing a decreasing height together with a negative z-speed (meaning ascending) ... the big difference in IMU & GPS velocities indicate the same. Let's see what we can pinpoint once the correct mobile device DAT log is available.

Thanks for your input, @slup. Since I sync'd before copying the FLY012.DAT info, does that mean that further analysis is not possible?

I appreciate everyone's interest and responses so far.

Best Regards,

BT

 

[RayOZ]

Did you sent it to a 3rd party repairer? If you sent it to DJI, and that it was faulty, DJI should replace it for free under warranty (if it's under 1 year). And DJI Care Refresh would've been cheaper.

 

[billt1970]

Did you sent it to a 3rd party repairer? If you sent it to DJI, and that it was faulty, DJI should replace it for free under warranty (if it's under 1 year). And DJI Care Refresh would've been cheaper.

Thanks, @RayOZ. Yes I took it to a local authorized repair center that I could drive to. My first repair at DJI two years ago took 6 weeks; my second repair at an authorized center in NJ was better, but still took 3 weeks. Baltimore Drone Service will have it back to me within 10 days. The repairs include a new camera and gimbal assembly, IMU mount and dampeners, and compass assembly. The M2P generates steady income for me, so I need an operational drone ASAP. Some day I will be able to afford two, with one as backup, but not for a while.

I do have insurance with droneinsurance.com and have put a claim in to them. We'll see what happens there.

Again, thanks for all the interest and comments/suggestions.

Best Regards,

BT

 

[billt1970]

Good Morning, Friends.

I think I MAY have found the additional flight information suggested by @sar104 and @slup, FLY013.DAT. Please let me know if this helps in the analysis.

Best Regards,

BT

 

[billt1970]

I have reviewed the flight documentation second by second from 231 seconds flight time to the end at 265 seconds. FWIW, here is my pilot's interpretation of events documented in the flight logs and records.

"I was flying a photo mission for a real estate agent client. The flight was normal until approximately 231 seconds into flight.

At that time I was positioning the aircraft (at altitude 398') for another photo, when it suddenly started moving laterally to the right although I was not attempting any movement with the flight controls. That movement stabilized briefly after approximately 12 seconds (at altitude 349'), at which time the aircraft began moving, independent of the flight controls, and rapidly descending to the left at between 21mph and 52mph!! This rapid lateral and descending movement ended at approximately 255 seconds into flight time (at altitude 65'), whereupon the aircraft came in contact with a nearby tree and fell downward through the foliage to the ground. I was able to hear the noise associated with the aircraft falling through the tree. The flight records then end at 265 seconds into flight.

The live view remained active at this time and I could see on the iPhone 6+ with the controller that it was indeed on the ground. I quickly walked to where I heard the "falling through tree" sounds about 20 yards away and found the aircraft sitting upright on the ground, with NO propellers!!! "

This is a direct copy of what I submitted to droneinsurance.com in my claim.

Any other comments, interpretations or suggestions are welcome.

Best Regards,

BT

 

[slup]

Good Morning, Friends.

I think I MAY have found the additional flight information suggested by @sar104 and @slup, FLY013.DAT. Please let me know if this helps in the analysis.

Best Regards,

BT

Unfortunately that is only a 72sec long power on/off sequence together with a movement of the AC, probably by hand, that will not give us more to go on ... the one from the flight ends with FLY012.DAT

 

[Slade]

I hope you have an Insurance policy for your drone.

 

[billt1970]

Unfortunately that is only a 72sec long power on/off sequence together with a movement of the AC, probably by hand, that will not give us more to go on ... the one from the flight ends with FLY012.DAT

Thanks, @slup. I suspected that wasn't the right file, but found it and wanted to throw it in here for comment. For the FLY012.DAT file, must that be copied after each flight AND before syncing, or if there are multiple flights between syncs, will each of the flight files be preserved?

BT

 

[billt1970]

I hope you have an Insurance policy for your drone.

Thanks @Slade. I have on demand insurance from droneinsurance.com. I filed the claim this morning; we'll see how they respond. Stay tuned.

BT

 

[sar104]

Thanks, @slup. I suspected that wasn't the right file, but found it and wanted to throw it in here for comment. For the FLY012.DAT file, must that be copied after each flight AND before syncing, or if there are multiple flights between syncs, will each of the flight files be preserved?

BT

The DJI GO 4 app doesn't auto-sync - you do that manually. But with the latest iOS version, which you are on, when it syncs flight records it deletes the associated DAT files from the mobile device.

 

[scoble08]

The DJI GO 4 app doesn't auto-sync - you do that manually. But with the latest iOS version, which you are on, when it syncs flight records it deletes the associated DAT files from the mobile device.

Is the IOS DAT file always named FLY012.DAT? What is the name of the Android (Smart Controller) .DAT file?

 

[billt1970]

The DJI GO 4 app doesn't auto-sync - you do that manually. But with the latest iOS version, which you are on, when it syncs flight records it deletes the associated DAT files from the mobile device.

My apologies if my last comment was obtuse. My question relates to the "associated .DAT files." Is the name ALWAYS FLY012.DAT, which is overwritten on the next flight, or does the numbering increment with each subsequent flight? Such as, FLY013.DAT, FLY014.DAT and on in sequence. I am trying to understand if I need to preserve the .DAT file after each flight, or preserve them all from flights between syncs? Does the question make sense now?

BT

 

[sar104]

Is the IOS DAT file always named FLY012.DAT? What is the name of the Android (Smart Controller) .DAT file?

My apologies if my last comment was obtuse. My question relates to the "associated .DAT files." Is the name ALWAYS FLY012.DAT, which is overwritten on the next flight, or does the numbering increment with each subsequent flight? Such as, FLY013.DAT, FLY014.DAT and on in sequence. I am trying to understand if I need to preserve the .DAT file after each flight, or preserve them all from flights between syncs? Does the question make sense now?

BT

The file number increments every time the aircraft is powered on. The number associated with any given flight is listed in the flight txt log as HOME.dataRecorderFileIndex.

Whether or not you want to retrieve and save them before synchronizing flight records is personal preference. I'd certainly recommend retrieving after any kind of incident where you might need to examine the data.

 

[scoble08]

The file number increments every time the aircraft is powered on. The number associated with any given flight is listed in the flight txt log as HOME.dataRecorderFileIndex.

Whether or not you want to retrieve and save them before synchronizing flight records is personal preference. I'd certainly recommend retrieving after any kind of incident where you might need to examine the data.

Thx.

 

[billt1970]

The file number increments every time the aircraft is powered on. The number associated with any given flight is listed in the flight txt log as HOME.dataRecorderFileIndex.

Whether or not you want to retrieve and save them before synchronizing flight records is personal preference. I'd certainly recommend retrieving after any kind of incident where you might need to examine the data.

Thanks, @sar104, that's what I needed to know. So, next time there is an "incident" I can make a conscious decision to save the .DAT file for the flight before syncing.

BT